【报告简介】   

Convection in global climate models (GCM) is determined by convective parameterization schemes, mainly through triggers and closures. Different types of triggers and closures exist in current parameterization schemes. Triggers are largely ad hoc, and have never been systematically evaluated. Closures vary from convective quasi-equilibrium, CAPE removal to moisture convergence and boundary layer control. As GCM resolution increases to the grey zone (~10 km), do we still expect current convective parameterization scheme to work well? In this talk, I will use observations and cloud-resolving model simulations of tropical and midlatitude convection to evaluate factors controlling convection. The relationship between coarse-grained variables and convection for a range of averaging domain sizes representing GCM horizontal resolutions will be evaluated. I will show that moisture convergence and CAPE (convective available potential energy) generation by tropospheric ascending motion are highly correlated with the activity of convection on all scales down to sub-10 km. On the other hand, CAPE and PBL turbulent kinetic energy (TKE)-based closures do not capture the variation of convection with the coarse-grained fields. The scale-awareness will be discussed in the context of convective parameterization as GCM resolution increases.